Our studies initially focused on patients with CD4 counts above 200 cells/mm3, administering IL-2 for 5 days approximately every 2 months at doses ranging from 6 to 18 million units/d. The courses of IL-2 were well-tolerated, although most of the patients required dosage reductions due to IL-2 related adverse effects. Sustained improvement in CD4 number was seen primarily in patients with greater than 200 CD4 cells/mm3. There also was a transient increase in viral load as measured by the bDNA assay seen at day 6 to day 8 following initiation of IL-2 therapy. Responses in CD4 number were less common in patients with lower baseline CD4 counts. Based on the preliminary results seen in our open trial, we undertook a 60 patient randomized trial to evaluate IL-2 therapy in patients with CD4 counts above 200 cells/mm in combination with currently approved antiretroviral therapies. The study opened in April 1993 and was completed in February of 1995, with 60 patients enrolling. This study also showed in a controlled setting that intermittent therapy with IL-2 can lead to a substantial and sustained increase in CD4 cell counts without leading to an increase in plasma viral load. More recently, we have focused on improving the tolerance of IL-2, by decreasing the dose and duration of therapy, and by evaluating alternative methods of administering IL-2. We had enrolled patients in an extension phase of ongoing studies to determine whether administration of corticosteroids with IL-2 can lead to improved tolerance of IL-2 without interfering with the immunomodulatory effects. This phase has been discontinued due to the occurrence of avascular necrosis of the hip in some patients receiving prednisone. We continue to follow patients receiving IL-2 to determine the long term side effects and immunologic activity of IL-2. In addition, in combination with labeling studies, we are investigating the mechanisms leading to the profound CD4 cell increases seen with intermittent IL-2 therapy. These studies are potentially important because they are the first ones to suggest that immunomodulating agents combined with antiretroviral agents may have a benefit in patients with HIV infection.